Lyria 3 Pro Preview
Lyria 3 Pro Preview is a multimodal model made by Google, added to the catalog on 30 Mar 2026. It has a context window of 1,048,576 tokens and has no per-token charge. It accepts text and images and produces text and audio. Typical uses: generating music from text inputs, exploring music generation with image inputs and producing text alongside audio.
Price checked ยท source: OpenRouter
What Lyria 3 Pro Preview is
Google Lyria 3 Pro Preview is a music-generation model in Google's Lyria 3 family, available through the Gemini API. It accepts text and image inputs and can produce both text and audio. The catalog identifies it as a model for music generation but does not specify which musical controls or formats are available. Its context window is 1,048,576 tokens, and its maximum output is 65,536 tokens.
The large context window provides room for extensive input, while the output limit sets the maximum response size. Text and image inputs paired with audio output make the model relevant to workflows involving written or visual material and generated music; text output is also supported. The listing does not explain how images affect generation, which audio formats or durations are returned, or what other generation controls are available. It lists support for response_format but does not name the formats.
Lyria 3 Pro Preview may suit experimentation with music generation from text or image inputs, as well as workflows that use both text and audio outputs. Its context capacity can accommodate lengthy supplied material, although that capacity alone does not establish a particular result or quality. The model is identified as a preview, and the available facts do not provide further detail about its generation behavior.
Key features
- Accepts text and image inputs
- Produces text and audio
- Has a 1,048,576-token context window
- Supports up to 65,536 output tokens
- Supports response_format
Typical uses
- Generating music from text inputs
- Exploring music generation with image inputs
- Producing text alongside audio
- Working with lengthy input material
Lyria 3 Pro Preview pricing
US dollars, list price through OpenRouter.
| Item | Standard |
|---|---|
| Input, per 1M tokens | $0 |
| Output, per 1M tokens | $0 |
No price change since it was added on 30 Mar 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0 |
| Summarise a 50-page PDF25,000 in / 800 out | $0 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0 |
| Read a 300-page contract set150,000 in / 1,500 out | $0 |
Lyria 3 Pro Preview specifications
| Model id | google/lyria-3-pro-preview |
|---|---|
| Version | lyria-3-pro-preview-20260330 |
| Context window | 1,048,576 tokensAbout 786,432 English words |
| Max output | 65,536 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 30 Mar 2026 |
| Input types | text, image |
|---|---|
| Output types | text, audio |
| Tool calling | No |
| Structured outputs | No |
| Reasoning effort control | No |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Lyria 3 Pro Preview yet.